Exercise 2 (20 minutes)

When you're done, assist some of the others ...

1. Regress poverty on our socialism index. With one more year of socialist history, poverty changes by: percentage points (two digits).

R solution -> dont' peak too early ;) !

# Estimate regression models.
ols_1 <- lm_robust(data = Dat, formula = poverty ~ socialist)
modelsummary(list("Poverty" = ols_1), # Nicely-formatted table.
             statistic = NULL, # Don't report stat. inference
             gof_map = c("nobs", "r.squared"))

	Poverty
(Intercept)	11.618
socialist	-0.103
Num.Obs.	145
R2	0.007

2. Regress poverty on civic and political citizenship rights. With a one unit-increase in citizenship rights, poverty changes by: percentage points (two digits).

R solution -> dont' peak too early ;) !

# Estimate regression models.
ols_2 <- lm_robust(data = Dat, formula = poverty ~ citizen_rights)
modelsummary(list("Poverty" = ols_2), # Nicely-formatted table.
             statistic = NULL, # Don't report stat. inference
             gof_map = c("nobs", "r.squared")) 

	Poverty
(Intercept)	27.127
citizen_rights	-2.154
Num.Obs.	145
R2	0.189

3. Which of the two is a better model of the variance of poverty across the world? The one with socialismThe one with citizenship rights, as the R2 showsThe one with citizenship rights, as the stronger beta coefficient shows

4. Make a scatterplot with civic and political citizenship rights on the Y-axis and socialism as predictor on the X-axis.

R solution -> dont' peak too early ;) !

ggplot(data = Dat, aes(y = citizen_rights, x = socialist)) +
  geom_text(aes(label = country)) +
  labs(x = "Our socialism index", 
       y = "Freedom House index of citizenship rights") +
  theme_minimal() +
  guides(color = "none")

5. Regress citizenship rights on socialism. With one more year of socialism, citizenship rights change by:

R solution -> dont' peak too early ;) !

# Estimate regression models.
ols_3 <- lm_robust(data = Dat, formula = citizen_rights ~ socialist)
modelsummary(list("Citiz. Rights" = ols_3), # Nicely-formatted table.
             statistic = NULL, # Don't report stat. inference
              gof_map = c("nobs", "r.squared")) 

	Citiz. Rights
(Intercept)	7.862
socialist	-0.052
Num.Obs.	145
R2	0.045

6. Make a regression table that displays all three models. Based on the table, discuss with your neighbor: What do the results suggest about whether there is a freedom-equality trade off?

R solution -> dont' peak too early ;) !

modelsummary(list("Poverty" = ols_1, "Poverty" = ols_2, "Citiz. Rights" = ols_3), # Nicely-formatted table.
              gof_map = c("nobs", "r.squared")) 

	Poverty	Poverty	Citiz. Rights
(Intercept)	11.618	27.127	7.862
	(1.674)	(4.290)	(0.331)
socialist	-0.103		-0.052
	(0.079)		(0.018)
citizen_rights		-2.154
		(0.425)
Num.Obs.	145	145	145
R2	0.007	0.189	0.045